Cue-exposure treatments (CETs) in which addicted individuals are exposed to cues associated with drug use, but prevented from responding, have demonstrated limited effectiveness. This is potentially due to the fact that extinction learning is highly context dependent-i.e. behavior extinguished in a context other than the one in which it was learned, is re-expressed or 'renewed' when the animal returns to the original learning context. The overarching goal of our collaborative research program is to increase the efficacy of CETs by incorporating contextual cues from the smoker's real-world smoking environment into treatment, thereby preventing renewal. However, in order to rationally develop novel CETs that incorporate personal smoking environment cues, it is first necessary to firmly establish that such cues result in reactivity distinct from standard smoking environment cues and traditionally used proximal cues (e.g. lit cigarette) and that they provoke smoking behavior (i.e. self- administration). In order to achieve these initial steps, seventy-eight (n=78) adult smokers will undergo BOLD fMRI scanning while they view picture cues depicting personal and experimenter generated standard smoking and nonsmoking environments and while viewing smoking and nonsmoking proximal cues. Personal environment cues will be acquired by training smokers to take pictures in real-world settings. In addition, the effects of personal smoking and nonsmoking environment cues on cigarette smoke self-administration and self-reported craving will be assessed in two lab sessions. We hypothesize that in addition to increased BOLD signal in cue-reactivity areas, exposure to personal smoking environment cues will result in greater activation of a drug-context association area (hippocampus) relative to proximal smoking cues; and greater activation of self-relevance processing areas (mPFC, PCC) relative to standard smoking environment cues. We further hypothesize that smokers will exhibit greater smoke self-administration (i.e. greater probability of smoking, decreased latency to smoke, greater puff volume) and craving in response to viewing personal smoking as compared to nonsmoking environment cues. Additionally, we will explore relations between behavioral and brain measures of connectivity during cue-exposure. If our hypotheses are supported, then the incorporation of personal smoking environments into CET is warranted and the rational development and evaluation of novel CETs can follow. In addition, the NIH is invested in the development of novel pharmacotherapies that enhance CET or disrupt reconsolidation of activated drug memories; the efficacy of these treatments can potentially be improved by incorporating more ecologically valid and personally relevant drug cues such as those proposed for study here. Finally, despite a rich preclinical literature and over 300 prior human cue-reactivity studies, very little is known regarding the environment as a determinant of smoking behavior in humans. As such, the proposed research, in addition to informing treatment development, will provide novel mechanism information regarding an important, yet understudied, determinant of drug taking and relapse.